dc7800_ has quit [Read error: Connection reset by peer]
antoinelassagne4 has joined #aarch64-laptops
antoinelassagne has quit [Ping timeout: 480 seconds]
antoinelassagne4 is now known as antoinelassagne
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
dc7800 has joined #aarch64-laptops
dc7800_ has quit [Ping timeout: 480 seconds]
tobhe_ has joined #aarch64-laptops
tobhe has quit [Ping timeout: 480 seconds]
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
<Arc[m]>
<JensGlathe[m]> "On my [discussion thread](https:..." <- So from a quick skim, seems like sound is still not working, and it still requires Windows presented to install Linux?
hexdump01 has joined #aarch64-laptops
hexdump0815 has quit [Ping timeout: 480 seconds]
<JensGlathe[m]>
You need the Windows partition to fetch the firmwares, yes. There are other ways, but for updates of firmware it‘s convenient. Sound shouldn’t be too hard (what an odd thing to say). X1 sound setup is well known and doesn’t vary much on these devices.
<JensGlathe[m]>
My impression was that it’s fully enabled, though. Can‘t verify, I don’t have the Omnibook 5.
<Arc[m]>
<JensGlathe[m]> "You need the Windows partition..." <- I haven't looked into the firmware blob side much but how about Fedora's way? Can it completely eliminate the Windows partition?
<Arc[m]>
Iirc Fedora can be boot without pulling fw from Windows' side.
dc7800 has joined #aarch64-laptops
dc7800_ has quit [Read error: Connection reset by peer]
<JensGlathe[m]>
You can boot and use the laptop without the fw, just not use all the features (like, sound). My recommendation would be to boot it up first, using the documented way. Before shooting for a setup with a different distro.
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
hexa- has quit [Quit: Disconnected]
hexa- has joined #aarch64-laptops
jhovold has joined #aarch64-laptops
dc7800_ has quit []
dc7800 has joined #aarch64-laptops
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Ping timeout: 480 seconds]
dc7800 has joined #aarch64-laptops
dc7800_ has quit [Read error: Connection reset by peer]
SpieringsAE_sff has quit [Ping timeout: 480 seconds]
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
pbrobinson has quit [Ping timeout: 480 seconds]
dc7800 has joined #aarch64-laptops
dc7800_ has quit [Read error: Connection reset by peer]
SpieringsAE_sff has joined #aarch64-laptops
<anthony25>
you wouldn't have gpu acceleration without the fw, which is more annoying than sound (you would use llvmpipe / lavapipe)
pbrobinson has joined #aarch64-laptops
ektor5 has quit [Ping timeout: 480 seconds]
ektor5 has joined #aarch64-laptops
ektor5 has quit [Ping timeout: 480 seconds]
ektor5 has joined #aarch64-laptops
pbrobinson has quit [Ping timeout: 480 seconds]
icecream95 has joined #aarch64-laptops
pbrobinson has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
dc7800 has joined #aarch64-laptops
<icecream95>
robclark: I'm wondering if there might be a bug in ir3_max_const...
<icecream95>
Rendering issues in Doom 3 BFG Edition go away with IR3_SHADER_DEBUG=nouboopt, and one of the shaders involved (depth_skinned.vertex) uses a UBO with a vec4[408] array
<icecream95>
Without nouboopt, skinned NPC meshes render correctly once I reduce the UBO size below about 310. I could try to write a standalone reproducer if that would be helpful
anarsoul has quit [Read error: Connection reset by peer]
anarsoul has quit [Read error: Connection reset by peer]
anarsoul has joined #aarch64-laptops
miklv has quit [Ping timeout: 480 seconds]
miklv_ is now known as miklv
notHorseface has joined #aarch64-laptops
<robclark>
hmm, looks like we use same value for (almost) everything.. which is probably not correct. Which device?
<icecream95>
robclark: X1-85
<robclark>
is it just vs+fs? Or what stages are active? And any idea about const usage in other stages?
<icecream95>
Just vs+fs, fragment shader is literally just output vec4(0,0,0,1), the only other uniform is vec4[4], and there are a few more constants used in the VS
<robclark>
ok
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
<robclark>
looks like LB size for hamoa/purwa should be same as a750.. presumably valve has tried doom3 bfg?
dc7800_ has quit [Ping timeout: 480 seconds]
dc7800 has joined #aarch64-laptops
<icecream95>
In case it helps, I uploaded an apitrace at https://drive.google.com/file/d/1EgivJLmZoZAB62CRPX4mor1D4suutvmx (ignore text rendering issues, that's an RBDOOM-3-BFG bug). For example, call 1121929 has the issue, though it only shows up [in the depth buffer] if you trace until call the following draw at 1121934
<robclark>
hmm, seems to be ok on glymur (unhelpfully).. I don't have a (working) hamoa thing handy.. you could try it w/ zink, it's possible there is some gallium specific bug I suppose?
<icecream95>
The bug reproduces under Zink and IR3_SHADER_DEBUG=nouboopt also fixes it there
<robclark>
hmm, ok
<robclark>
maybe cwabbott could try the trace on a750? It looks like vec4[408] on it's own should be ok. But tracking and implementing relevant quirks for the various GPUs is still wip..
<robclark>
Ig I can try dump/diff images compared to nouboopt just to rule out that I'm not noticing the corruption (not sure how obvious it should be?)
<icecream95>
Pretty obvious... big black triangles going across half of the screen
<robclark>
hmm, not seeing that.. apitrace dump-images slows it down enough to get a better look
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
<notHorseface>
but from what I can see on the packages page of archlinuxarm, those packages ARE all up-to-date...
<notHorseface>
thank you for that, any way.
<notHorseface>
lol
dc7800 has quit [Ping timeout: 480 seconds]
dc7800 has joined #aarch64-laptops
SpieringsAE has joined #aarch64-laptops
SpieringsAE has quit [Quit: SpieringsAE]
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
<f_>
Arch Linux ARM looks un(der)maintained to me
<f_>
while I heard good things about Arch Linux Ports
<HdkR>
Looking forward to official ARM64 support from the Arch team :)
dc7800 has joined #aarch64-laptops
dc7800_ has quit [Ping timeout: 480 seconds]
dc7800_ has joined #aarch64-laptops
dc7800 has quit [Read error: Connection reset by peer]
dc7800 has joined #aarch64-laptops
dc7800_ has quit [Read error: Connection reset by peer]
SpieringsAE_sff has quit [Read error: Connection reset by peer]
SpieringsAE_sff has joined #aarch64-laptops
<Jadesheher[m]>
has anyone ever gotten spotify to work in Firefox?
<Jadesheher[m]>
i tried the asahi widevine installer and its showing up but i cant play anything
<Jadesheher[m]>
or has anyone ever used any emulation thing to run the normal app?
<steev>
Jadesheher[m]: you could just install google chrome (unless you wanna avoid that) which has properly working widevine (i can at least watch netflix in it, dunno about spotify)
<Jadesheher[m]>
hmm not a fan of chrome will chromium work?
<Jadesheher[m]>
can i make it just launch the spotify window? without the rest of it
<Derek[m]>
Jadesheher[m]: Any chrome based browser should work. Spotify works for me in vivaldi.
<Jadesheher[m]>
do i need a special chromium version?
<Jadesheher[m]>
i currently have the ones from archlinux ports repo
<Derek[m]>
I just asked claude to make it work in firefox. And it totally does but idk if I could replicate it.
<steev>
just needs widevine properly set up, and the only browser i've used that has done so is google chrome. alternatively, use other media sources that don't require drm
<Jadesheher[m]>
steev: i would but some artists i listen to only upload to spotify
<Jadesheher[m]>
and youtube music but thats about as performant as a potatoe
<Jadesheher[m]>
it runs crappy on my phone and in the browser
SpieringsAE_sff has quit [Remote host closed the connection]
SpieringsAE_sff has joined #aarch64-laptops
<Jadesheher[m]>
for fucks sake it was mpv that was stopping the playback
<Jadesheher[m]>
i just tried and it works
<Jadesheher[m]>
fuuuuck i spent 2 hours trying to get the app to run...
SpieringsAE_sff has quit [Ping timeout: 480 seconds]
<deathmist>
HdkR: oh btw I realized on the X1E Vivobook I actually don't have the problem under WSL (battery charge bounced between 78.4% and 79.5% with some periodic downclocking to 3.2 GHz for 30/45 seconds), so just native Linux result would be enough. after tweaking my Linux side limit to 79/80 I also seemingly got it to stay at ~79.5% now
<deathmist>
will definitely continue with more testing with other all-core compiles or similar and seeing if it still keeps charging past the set limit now. too bad I at least don't seem to have bypass charging here, hope the A16 is different
<valpackett>
mpv could stop firefox's playback? o.0
<deathmist>
pick your poison really, I didn't even bother looking at the former on my laptop based on my last experiences with its ancient glibc some years back etc
<notHorseface>
oh I see.
<notHorseface>
glibc
<notHorseface>
aarch64coreglibc2.43+r22+g8362e8ce10b2-2GNU C Library